以人臉照片辨識不良寵物飼主之研究

以作者查詢圖書館館藏

、以作者查詢臺灣博碩士

、以作者查詢全國書目

、勘誤回報

、線上人數：29

、訪客IP：18.226.98.89

姓名

趙千儀(Chien-Yi Chao) 查詢紙本館藏

畢業系所

會計研究所企業資源規劃會計碩士在職專班

論文名稱

以人臉照片辨識不良寵物飼主之研究
(Identifying bad pet owners from pictures with CNN models)

相關論文

★ ERP樣版導入方法論之個案研究	★ Talent Management Using Characteristics Model of RPG
★ 以情境理論探討顧客在保固期後再回原廠消費之意願─以B公司為例	★ 企業接班模式之探討—以G公司為例
★ 影響消費者透過行動美妝軟體購買美妝品之關鍵因素研究－使用AHP方法	★ 客製化產品成本導入ERP之計算架構-以A公司製造自動化設備為例
★ 應用中介系統改善企業流程-以S公司為例	★ 結合群眾智慧與基本面應用於上市公司投資策略之研究
★ 共享服務中心外包衍伸問題之探討-以M跟S集團在台分公司為例	★ 保險業於大數據時代導入 SAP HANA 之關鍵成功因素 - 以 N公司為例
★ 以八階段分析法進行談判策略之個案研究	★ 以AHP法研究企業評選顧問公司之準則
★ B2B拍賣網上結標價和結標時間的預測	★ 應用啟發式演算法挖掘與表示商業流程的碳排量屬性
★ ERP為基準碳足跡計算之觀念性架構發展	★ 作業基礎制碳足跡方法-以作業制成本法為基礎發展產品碳足跡計算之方法

檔案

[Endnote RIS 格式]

[Bibtex 格式]

[相關文章]

[文章引用]

[完整記錄]

[館藏目錄]

至系統瀏覽論文 (2027-6-15以後開放)

摘要(中)

隨著人口結構和社會經濟型態的改變，越來越多人選擇飼養寵物作為生活伴侶，擁有寵物也已經被證實具有許多的好處。然而，雖然很多人都知道擁有寵物可以帶來許多好處，但寵物遺棄、虐待動物和不當飼養事件至今還是相當普遍。而絕大多數飼主其實就是潛在的棄養者、施虐者，我們無法知道哪一位飼主在未來某一天會棄養，甚至虐待自己的寵物，若是能將潛在不良飼主第一時間排除在外，在第一步就做好預防，進而能有效抑制一連串問題的發生。

因此，本研究嘗試透過社群媒體上蒐集黑名單白名單人臉照片，利用深度學習中的卷積神經網路（CNN）演算法進行寵物飼主的好壞辨識。本文的人臉識別CNN 模型結構，由六個卷積層和四個池化層，以及全連接層中的兩個隱藏層組成，我們使用了 Data augmentation（數據增強）來解決數據集不足的問題，使用Batch Normalization（批次正規化）克服模型難以訓練的問題，使用 Dropout 等方法減緩過擬合，並且使用 Adam 優化器和 Softmax 分類器進行人臉識別可以使訓練更穩定、更快收斂，有效提高準確率。

透過卷積神經網路很強的特徵提取圖像辨識的方式，可以快速有效地對寵物飼養人進行辨別。實驗結果表明，本研究 CNN 模型在寵物飼主人臉照片上的識別率為 80.09%。

摘要(英)

Along with the change in population structure and the social-economic pattern, adopting pets have become popular. However, as raising pets requires long term devotion and commitment, some pet owners may change their minds or may resort to abusing pets to release their emotion. To preventing pets from becoming the victims of bad owners, how to deter the potential abusers to adopt pets has become an important issue. As people with unstable emotion tend to have certain types of facial expression, this study proposed to build up discriminant model for bad owners by analyzing pictures. Even though analyzing facial images with Nero networks have been studied in many areas, to the best of our knowledge, no one has applied related knowledge to this issue.

The proposed model utilizes CNN with six Convolution layers, four Pooling layers, and two Fully Connected Layer at the end. As the number of pictures was not sufficient, Data augmentation was utilized to increase the data size. Batch Normalization is utilized to quickly converge the model parameters as the number of data are still relative limited. Dropout and regularization methods are also adopted to relieved the issue of overfitting. Numerous hyper parameters tuning were attempted, and result showed that the accuracy can reach 80.09%.

關鍵字(中)

★ 寵物棄養
★ 虐待動物
★ 不當飼養
★ 飼主
★ 深度學習
★ 人臉辨識
★ 卷積神經網路

關鍵字(英)

★ Pet abandon
★ Pet abuse
★ Improper feeding
★ Owner of the pet
★ Deep learning
★ Convolutional Neural Network
★ CNN

論文目次

中文摘要 .................................................................................i
Abstract..................................................................................ii
誌謝......................................................................................iii
目錄......................................................................................v
圖目錄 ...................................................................................vii
表目錄 ...................................................................................viii
第一章緒論 ..............................................................................1
1-1 研究背景與動機....................................................................1
1-2 研究目的 .........................................................................7
1-3 研究架構 .........................................................................7
第二章文獻探討 ..........................................................................8
2-1 棄養、虐待及不當飼養行為之意向 ...................................................8
2-1-1 棄養動物行為 ...............................................................8
2-1-2 虐待動物行為 ...............................................................9
2-1-3 不當飼養行為 ...............................................................10
2-2 深度學習卷積神經網路在人臉識別之研究 ............................................ 12
第三章研究方法 ..........................................................................15
3-1 研究流程 .........................................................................15
3-2 卷積神經網路......................................................................16
3-2-1 卷積層 .....................................................................17
3-2-2 池化層 .....................................................................18
3-2-3 全連接層 ...................................................................19
3-2-4 激活函數 ...................................................................20
第四章研究實驗分析及結果 ................................................................21
4-1 資料蒐集及預處理..................................................................21
4-2 研究工具 .........................................................................23
4-3 研究過程與分析....................................................................24
4-3-1 第一個實驗階段..............................................................26
4-3-2 第二個實驗階段..............................................................28
4-3-3 第三個實驗階段..............................................................30
4-3-4 第四個實驗階段..............................................................32
4-3-5 第五個實驗階段..............................................................34
4-3-6 第六個實驗階段..............................................................36
4-4 模型評估及預測....................................................................38
第五章結論 ..............................................................................40
5-1 研究結論 .........................................................................40
5-2 研究限制與未來建議................................................................41
5-2-1 研究限制 ...................................................................41
5-2-2 未來建議 ...................................................................42
第六章參考文獻 ..........................................................................43

參考文獻

中文文獻
[1] 丁春木.(2007).上帝的靈和約與動物的關係: 從創世紀 7: 15; 9: 8~ 11 探討建構生態公義的神學基礎.
[2] 王嘉顺.(2011).宠物饲养者的社会心理行为分析.医学与社会, 24,3,85-87.
[3] 行政院農業委員會.(2013).近年來臺灣寵物產業發展情形及相關管理措施.
[4] 吳怡伶.(2006).動物溝通: 寵物作為人際互動的中介.世新大學,口語傳播學系, 碩士學位論文.
[5] 張嘉妤,張曇欣,朱珮馨,陳慧心,張智涵.(2014).台灣流浪動物現況探討.
[6] 楊皖淩.(2021).優化 CNN 建立聲音滿意度辨識模型.國立中央大學企業管理學系.
[7] 羅淑芬,孔秀美,曾美蓮,林淑娟,張秀香.(2003).寵物治療在護理專業領域之臨床應用.護理雜誌,50,1,93-97.

英文文獻
[8] Alleyne, E., Tilston, L., Parfitt, C. and Butcher, R. Adult-perpetrated animal abuse: Development of a proclivity scale. Psychology, Crime & Law, 21, 6 (2015), 570-588.
[9] Arel, I., Rose, D. C. and Karnowski, T. P. Deep machine learning-a new frontier in artificial intelligence research [research frontier]. IEEE computational intelligence magazine, 5, 4 (2010), 13-18.
[10] Ascione, F. R. Children who are cruel to animals: A review of research and implications for developmental psychopathology. Anthrozoös, 6, 4 (1993), 226-247.
[11] Bashar, A. Survey on evolving deep learning neural network architectures. Journal of Artificial Intelligence, 1, 02 (2019), 73-82.
[12] Beck, A. M. and Meyers, N. M. Health enhancement and companion animal ownership. Annual review of public health, 17, 1 (1996), 247-257.
[13] Boureau, Y.-L., Ponce, J. and LeCun, Y. A theoretical analysis of feature pooling in visual recognition. City, 2010.
[14] Dagui, L. TensorFlow+ Keras deep learning artificial intelligence practice application [M]. Tsinghua University Press Beijing, City, 2018.
[15] Dhall, A. and Hoey, J. First impressions-predicting user personality from twitter profile images. Springer, City, 2016.
[16] Ding, X., Zhang, X., Zhou, Y., Han, J., Ding, G. and Sun, J. Scaling up your kernels to 31x31: Revisiting large kernel design in cnns. arXiv preprint arXiv:2203.06717 (2022).
[17] Eduard, A. and Shashkin, A. Journal of Physics: Conference Series (2019).
[18] Fatjó, J., Bowen, J., García, E., Calvo, P., Rueda, S., Amblás, S. and Lalanza, J. F. Epidemiology of dog and cat abandonment in Spain (2008–2013). Animals, 5, 2 (2015), 426-441.
[19] Fawcett, A. Why do cats enter shelters. Shelter Research, 6 (2008), 1-3.
[20] Flynn, C. Acknowledging the" Zoological connection": A sociological analysis of animal cruelty. Society & Animals, 9, 1 (2001), 71-87.
[21] Friedmann, E., Katcher, A. H., Lynch, J. J. and Thomas, S. A. Animal companions and one-year survival of patients after discharge from a coronary care unit. Public health reports, 95, 4 (1980), 307.
[22] Gershman, K. A., Sacks, J. J. and Wright, J. C. Which dogs bite? A case-control study of risk factors. Pediatrics, 93, 6 (1994), 913-917.
[23] Gullone, E. Animal cruelty, antisocial behaviour, and aggression: More than a link. Springer, 2012.
[24] Hariri, W. Efficient masked face recognition method during the covid-19 pandemic. Signal, image and video processing (2021), 1-8.
[25] Jing, C., Song, T., Zhuang, L., Liu, G., Wang, L. and Liu, K. A survey of face recognition technology based on deep convolutional neural networks. Computer applications and software, 35, 01 (2018), 223-231.
[26] Johnson, S. A. Animal cruelty, pet abuse & violence: the missed dangerous connection. Forensic Research & Criminology International Journal, 6, 6 (2018), 403-415.
[27] Jucheng, Y., Na, L., Shanshan, F. and Ying, X. Review of face recognition methods based on deep learning [J]. Journal of Tianjin University of Science and Technology, 31, 06 (2016), 1-10.
[28] Jules-Macquet, R. Link between animal cruelty and human abuse: a review of the literature. De Rebus, 2014, 542 (2014), 26-28.
[29] Kellert, S. R. and Felthous, A. R. Childhood cruelty toward animals among criminals and noncriminals. Human relations, 38, 12 (1985), 1113-1129.
[30] Kim, G., Choi, I., Li, Q. and Kim, J. A CNN-Based Advertisement Recommendation through Real-Time User Face Recognition. Applied Sciences, 11, 20 (2021), 9705.
[31] Lasri, I., Solh, A. R. and El Belkacemi, M. Facial emotion recognition of students using convolutional neural network. IEEE, City, 2019.
[32] Lee, H., Grosse, R., Ranganath, R. and Ng, A. Y. Convolutional deep belief networks for scalable unsupervised learning of hierarchical representations. City, 2009.
[33] Lu, P., Song, B. and Xu, L. Human face recognition based on convolutional neural network and augmented dataset. Systems Science & Control Engineering, 9, sup2 (2021), 29-37.
[34] Marston, L. C., Bennett, P. C. and Coleman, G. J. What happens to shelter dogs? An analysis of data for 1 year from three Australian shelters. Journal of Applied Animal Welfare Science, 7, 1 (2004), 27-47.
[35] Marston, L., Bennett, P. and Toukhasati, S. Cat admissions to Melbourne Shelters: A Report to the Bureau of Animal Welfare, December 2006. Melbourne. Animal Welfare Science Centre, Monash University (2006).
[36] Newberry, M. Associations between different motivations for animal cruelty, methods of animal cruelty and facets of impulsivity. Psychology, Crime & Law, 24, 5 (2018), 500-526.
[37] Parfitt, C. H. and Alleyne, E. Animal abuse proclivity: Behavioral, personality and regulatory factors associated with varying levels of severity. Psychology, Crime & Law, 24, 5 (2018), 538-557.
[38] Richard, C. and Reese, L. A. The interpersonal context of human/nonhuman animal violence. Anthrozoös, 32, 1 (2019), 65-87.
[39] SAI, E. C., HUSSAIN, S. A., KHAJA, S. and SHYAM, A. Student Attendance Monitoring System Using Face Recognition. Available at SSRN 3851056 (2021).
[40] Saranya, G., Sarkar, D., Ghosh, S., Basu, L., Kumaran, K. and Ananthi, N. Face Mask Detection using CNN. IEEE, City, 2021.
[41] Seksel, K. The behavioural manifestations of animal cruelty/abuse. City, 2004.
[42] Serpell, J. Beneficial effects of pet ownership on some aspects of human health and behaviour. Journal of the royal society of medicine, 84, 12 (1991), 717-720.
[43] Setialana, P., Jati, H., Wardani, R., Indrihapsari, Y. and Norwawi, N. M. Intelligent Attendance System with Face Recognition using the Deep Convolutional Neural Network Method. IOP Publishing, City, 2021.
[44] Silver, D., Huang, A., Maddison, C. J., Guez, A., Sifre, L., Van Den Driessche, G., Schrittwieser, J., Antonoglou, I., Panneershelvam, V. and Lanctot, M. Mastering the game of Go with deep neural networks and tree search. nature, 529, 7587 (2016), 484-489.
[45] Sun, A., Zhao, Y., Zhuang, G. and Zhang, G. Pets are irresponsible for COVID-19.
Authorea Preprints (2020).
[46] Tallichet, S. E. and Hensley, C. Rural and urban differences in the commission of animal
cruelty. International Journal of Offender Therapy and Comparative Criminology, 49, 6
(2005), 711-726.
[47] Vaughn, M. G., Fu, Q., DeLisi, M., Beaver, K. M., Perron, B. E., Terrell, K. and Howard, M. O. Correlates of cruelty to animals in the United States: Results from the National Epidemiologic Survey on Alcohol and Related Conditions. Journal of Psychiatric Research, 43, 15 (2009), 1213-1218.
[48] Ventura, C., Masip, D. and Lapedriza, A. Interpreting cnn models for apparent personality trait regression. City, 2017.
[49] Wirth, H. J. Abandoned Animals in Australia—Not Just Dumped Doggies and Cast Away Kittens. Australian Animal Welfare: St Leonards, Austrilia (2016).
[50] Xing, J., Fang, G., Zhong, J. and Li, J. Application of face recognition based on CNN in fatigue driving detection. City, 2019.
[51] Zhang, H., Qu, Z., Yuan, L. and Li, G. A face recognition method based on LBP feature for CNN. IEEE, City, 2017.
[52] Zheng, H. H. and Zu, Y. X. A normalized light CNN for face recognition. IOP Publishing, City, 2018.
[53] Zhang, T., Qin, R.-Z., Dong, Q.-L., Gao, W., Xu, H.-R. and Hu, Z.-Y. Physiognomy: Personality traits prediction by learning. International Journal of Automation and Computing, 14, 4 (2017), 386-395.

書籍部份
[54] 陳永維 and 李厚均選擇的能力 - 探索人工智慧的核心，一版. 大碩教育股份有限公司, 台北市. 網路來源
[55] AVMA. (2003).Veterinarians team up with plastic surgeons for dog bite prevention week. Retrieved from https://www.avma.org/javma-news/2003-05-15/veterinarians-team-plastic-surgeonsdog-bite-prevention-week-may-15-2003
[56] Benedikt Droste.(2021).Google Colab Pro+: Is it worth $49.99? .Retrieved from https://towardsdatascience.com/google-colab-pro-is-it-worth-49-99-c542770b8e5648
[57] Cinnamon AI Taiwan. (2019). CNN 入門-Overfitting. Retrieved from https://cinnamonaitaiwan.medium.com/cnn%E5%85%A5%E9%96%80-overfitting-d10acd15ec21
[58] Cinnamon AI Taiwan. (2019). 深度學習：CNN 原理. Retrieved from https://cinnamonaitaiwan.medium.com/%E6%B7%B1%E5%BA%A6%E5%AD%B8%E7%BF%92-
cnn%E5%8E%9F%E7%90%86-keras%E5%AF%A6%E7%8F%BE-432fd9ea4935
[59] Clay-Technology World. (2019). [Machine Learning] ReLU 函式介紹與程式實作. Retrieved from https://clay-atlas.com/blog/2019/10/20/machine-learning-chinese-relu-function/
[60] Clay-Technology World. (2019). [Machine Learning] Sigmoid 函數介紹與程式實作. Retrieved from https://clay-atlas.com/blog/2019/10/19/machine-learning-chinese-sigmoid-function/
[61] Clay-Technology World. (2019). [Machine Learning] Tanh 函式介紹與程式實作. Retrieved from https://clay-atlas.com/blog/2019/10/22/machine-learning-notes-tanh-function/
[62] CTWANT.(2021).一年發生 2 千件虐待動物案！高虹安籲設動保警察. Retrieved from https://www.ctwant.com/article/136033
[63] CTWANT.(2021).浪犬傷人 2／流浪犬不減反增揭動檢員少得可憐. Retrieved from https://www.ctwant.com/article/139357
[64] DPG 動物友善.(2021).台灣十大動保新聞 (下) 虐待動物案件不止、衝突未解. Retrieved from https://animal-friendly.co/pet/32240/
[65] ETtoday 寵物雲.(2021).整理包／受虐貓茶茶搶救不治！蘆洲男「熱水燙貓」始末一次看. Retrieved from https://pets.ettoday.net/news/2067089
[66] ETtoday 寵物雲.(2022).整理包／意外闖禍險終身監禁「比特犬索爾迎溫暖新家」始末曝光. Retrieved from https://pets.ettoday.net/news/2200194
[67] GETIT01. (2016). 人臉識別：回顧與展望. Retrieved from https://www.getit01.com/p2018041627642341/
[68] GGWithRabitLIFE . (2018). [機器學習 ML NOTE]Overfitting 過度學習. Retrieved from https://medium.com/%E9%9B%9E%E9%9B%9E%E8%88%87%E5%85%94%E5%85%94%E7%9A%84%E5%B7%A5%E7%A8%8B%E4%B8%96%E7%95%8C/%E6%A9%9F%E5%99%A8%E5%AD%B8%E7%BF%92-ml-note-overfitting-%E9%81%8E%E5%BA%A6%E5%AD%B8%E7%BF%92-6196902481bb
[69] Google Research. Welcome To Colaboratory. Retrieved from https://research.google.com/colaboratory/faq.html#resourcelimits%20%EF%BC%8CWelcome%20To%20Colaboratory
[70] IT 人忽逢桃林. (2020). AI 面試題之深入淺出卷積網路的平移不變性. Retrieved from https://iter01.com/517678.html
[71] LINE TODAY. (2021). 認養貓好難？中途媽列出超狂門檻. Retrieved from https://today.line.me/tw/v2/article/Kn56JR
[72] Model Whale. (2020). 数据增强方法综述. Retrieved from https://www.heywhale.com/mw/project/5e5f544eb8dfce002d7f3b53
[73] Nex3Z. (2017). Sigmoid 函数和 Softmax 函数的区别和关系. Retrieved from https://blog.nex3z.com/2017/05/02/sigmoid-%E5%87%BD%E6%95%B0%E5%92%8C-softmax-
%E5%87%BD%E6%95%B0%E7%9A%84%E5%8C%BA%E5%88%AB%E5%92%8C%E5%85%B3%E7%B3%BB/
[74] StackExchange. (2018). How to know if model is overfitting or underfitting? . Retrieved from https://stats.stackexchange.com/questions/355774/how-to-know-if-model-is-overfitting-orunderfitting
[75] Taiwan SPCA 台灣防止虐待動物協會. (2019). 改變的時刻到了 It′s Time for Change. City, 2019. Retrieved from https://www.spca.org.tw/work/services
[76] Taiwan SPCA 台灣防止虐待動物協會. (2021). 不當飼養. Retrieved from https://spca.org.tw/work/investigations/report/improper-rearing
[77] 大大通 - 君莫笑. (2021). Python 深度學習 1：Google Colab 介紹. Retrieved from https://www.wpgdadatong.com/tw/blog/detail?BID=B3230
[78] 中央通訊社. (2022). 疫情中領養寵物作伴經濟重啟後照顧問題頭大. Retrieved from https://www.cna.com.tw/news/aopl/202201290030.aspx
[79] 今周刊. (2021). 虐待動物等同刑事案件設「動保警察」拯救毛小孩. Retrieved from https://www.businesstoday.com.tw/article/category/183027/post/202101200015/
[80] 台灣動物社會研究會. (2021). 呼籲農委會正視「不當飼養犬隻」造成的「人犬衝突與傷害」要求農委會儘速公告飼養犬貓「飼主責任指引」杜絕不當飼養防止犬傷
人悲劇再度發生. Retrieved from https://www.east.org.tw/action/8571
[81] 行政院農業委員會. (2008). 愛護動物勿棄養養牠、愛牠，不要拋棄牠. Retrieved from https://www.coa.gov.tw/theme_data.php?theme=news&sub_theme=agri&id=3233
[82] 法律白話文運動. (2016). 潘宏朋｜你知道毛小孩在台灣法律上是什麼地位嗎？. Retrieved from https://plainlaw.me/2016/12/21/animal_protection_act/
[83] 動物保護資訊網. (2013).「102 年度動物保護公共政策研析專業服務計劃」動保法騷擾、虐待或傷害定義研究案成果報告. Retrieved from
https://animal.coa.gov.tw/download/resources/06_resources_a00_102_1.pdf
[84] 章志剛. (2010). 探討人格特質對寵物棄養意向之影響。真理大學企業管理學系碩士班碩士論文. Retrieved from https://hdl.handle.net/11296/hd38rs
[85] 陳祈安. (2021). 基於深度學習與本土大數據之人臉辨識 - 國立陽明交通大學電信工程研究所碩士論文. Retrieved from https://hdl.handle.net/11296/7m888h
[86] 搖滾尾巴. 不漠視「不當飼養」檢舉不當飼養該怎麼做. Retrieved from https://www.rocktail.com.tw/blog/view/331
[87] 維基百科，自由的百科全書. 臉部辨識系統. (2019). Retrieved from https://zh.wikipedia.org/zhtw/%E8%87%89%E9%83%A8%E8%BE%A8%E8%AD%98%E7%B3%BB%E7%B5%B1
[88] 網路溫度計. (2021). 茶茶不痛了...被潑熱水苦撐 11 天離世網提「虐待動物加重刑責」吸引上萬人附議. Retrieved from https://dailyview.tw/Popular/Detail/11592
[89] 機器之心. (2022). 大到 31x31 的超大卷积核，涨点又高效，一作解读 RepLKNet.Retrieved from https://www.jiqizhixin.com/articles/2022-03-17-8
[90] 環境資訊中心. (2021). 疫情催生日本寵物飼養熱潮動保團體憂棄養. Retrieved from https://e-info.org.tw/node/229615
[91] 聯合新聞網. (2021). 拜登迎白宮新成員第一狗狗「司令」. Retrieved from https://udn.com/news/story/6813/5977752

指導教授

許秉瑜(Ping-Yu Hsu)

審核日期

2022-6-21

推文